空间机器人神经网络自适应滑模目标操控轨迹跟踪控制  

Neural-network-based Adaptive Sliding-mode Trajectory Tracking Control of Space Robot Manipulating Space Target

作　　者：王嵩 王蜀泉[1] 张龙[1] WANG Song;WANG Shuquan;ZHANG Long(Key Laboratory of Space Utilization,Technology and Engineering Center for Space Utilization,Chinese Academy of Sciences,Beijing 100094,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院太空应用重点实验室,中国科学院空间应用工程与技术中心,北京100094 [2]中国科学院大学,北京100049

出　　处：《宇航学报》2023年第2期254-265,共12页Journal of Astronautics

基　　金：国家自然科学基金(61903354)。

摘　　要：针对参数未知的空间目标操控问题,考虑空间机器人负载不确定性、系统动力学不确定性和环境扰动等因素,为实现操作过程的稳定控制及机器人轨迹的有效跟踪,提出一种基于径向基神经网络估计不确定项的自适应增益非奇异终端滑模变结构控制器。首先基于拉格朗日法建立空间机器人的刚体动力学模型。考虑空间机器人基座姿态主动控制模式,使用径向基神经网络对模型中的不确定项进行估计。进而提出基于神经网络估计的非奇异终端滑模控制器,并针对不确定性和扰动的估计误差设计自适应增益,以期实现空间机器人系统轨迹跟踪控制的收敛。仿真校验结果表明所设计的控制方法具有较好的误差收敛速度和控制精度。An adaptive gain nonsingular terminal sliding mode controller based on radial basis function(RBF)neural network approximation is proposed to achieve the trajectory tracking control of the space robot considering the uncertain load,dynamical uncertainty and disturbance.Firstly,the rigid-body dynamics of space robot is established based on Lagrange’s method.The active attitude control mode of the base is chosen and RBF neural network is used to estimate the uncertainties in the model.Thereafter,a non-singular terminal sliding mode controller based on neural network estimation is proposed,and adaptive gain is designed for the unknown upper bound of uncertainty and disturbance,in order to achieve the convergence of trajectory tracking control of space robot system.The numerical simulation results show that the proposed controller has better convergence speed and better control precision.

关 键 词：空间机器人 非合作目标 滑模控制 神经网络自适应 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]